Modern vehicles include a vast array of sensors, such as air bag sensors, tire pressure sensors, engine sensors, seat belt sensors, and many others. The engine sensors, for example, can provide data about the vehicle's engine cycle to an engine control unit (ECU). Based on the data from the engine sensors, the ECU can then make changes to the engine cycle (e.g., spark plug timing and/or fuel injection parameters) to optimize engine performance.
As the number of vehicular sensors increases, integration is becoming a serious challenge for automakers. For example, wires connecting an ECU to its corresponding engine sensors can be several meters long. These wires are a significant cost factor in automotive systems and contribute to the overall weight of the vehicle. Because of this, some conventional wiring interfaces, such as PSI5 and DSI for example, limit the number of wires to two lines. These interfaces can be used in automotive airbag sensors, power train systems, and non-standardized pulse width modulated (PWM) protocols of ABS wheel speed sensors, among others. Unfortunately, absent countermeasures, the inductances and capacitances of these wires can give rise to noise at a resonant frequency set by the inductances and capacitances of the wires.
In some conventional implementations, an RLC filter (which includes a resistor and an optional by-pass inductor in parallel with the resistor) may be arranged between an ECU and its corresponding sensors. This RLC filter can be designed to attenuate noise at the resonant frequency, thereby helping to improve the performance of the vehicular sensing system. Unfortunately, however, the resistor and inductor included in such an RLC filter consume power, add to the cost of the sensor system, and may cause a voltage drop thereover. In addition, because different vehicles may include different sensors and different lengths of wires (i.e., different inductances and capacitances), it is difficult to set a single RLC filter to adequately attenuate the different resonant frequencies corresponding to the different vehicles in conventional sensing systems.
Therefore, conventional approaches are less than ideal for several reasons, and the inventors have devised improved sensor interfaces as described further herein.